Microsoft's New Fuel Cell Partner is Ready to Blow Away the Bloom Box
June 26, 2014 2:15 PM
comment(s) - last by
(Source: Redox Power)
Microsoft plans to test the fuel cells as a means of providing more efficient power to data centers
Microsoft Corp. (
Redox Power Systems LLC
, and the
University of Maryland
have been awarded a $5M USD government grant to test fuel cell technology in Microsoft's data centers. The grant comes courtesy of the
U.S. Department of Energy
Advanced Research Projects Agency – Energy (ARPA-E)
group. The grant is from ARPA-E's
Reliable Electricity Based on Electrochemical Systems (REBELS)
program, which is looking to fund novel on-site power generation technologies.
I. A Tough Market
Redox Power is one of the very few companies to have a large-size, kilowatt scale fuel cell product ready for commercialization. And given the market caps of publicly traded competitors or valuations of those privately held, that alone could make Redox Power billions.
But what is so exciting about Redox Power is that in a market whos current best case is "almost" paying off cost premium versus traditional mechanical fossil fuel generators, this startup may have the world's first commercial scale fuel cell to cost less than a traditional mechanical fossil generator over its lifetime.
A Redox Power PowerSERG2-80 (25 kW) system awaits deployment.
When it come to fuel cells, the technologies are real, but in most cases are either so hard to produce or so expensive that they wouldn't survive the rigors of the for-profit market. However, some kinds of fuel cells -- particularly
small backup power cells for personal electronics
-- have started to approach commercial readiness.
The market for large fuel cells has been rockier. According to sources, Germany's Siemens AG (
) -- a top fuel cell pioneer -- has dramatically scaled back its large scale development, General Electric Comp. (
) has completely abandoned the technology according to
a 2010 note
from the Gerson Lehrman Group (GLG). Thus far commercial successes are few and far between.
Sprint Corp. (
a mildly successful in-house large-scale hydrogen fuel cell project. But in terms of companies actually producing large quantities of mid-to-large size fuel cells,
Bloom Energy and its mildly successful "Bloom Energy Servers"
is among the few
commercial success stories
Currently America's large commercial fuel cell producers are:
Plug Power Inc. (
) (market cap: $780M USD)
FuelCell Energy Inc. (
) (market cap: $611M USD)
Ballard Power Systems Inc (
) (market cap: $540M USD)
Hydrogenics Corp. (
) (market cap: $173M USD)
All of those fuel cell startups -- targeting the backup power market -- lost money in 2013, despite generating substantial revenue.
Bloom Energy's blocky fuel cell "boxes" have seen modest sales. [Image Source: The Chronicle]
AT&T, Inc. (
), eBay, Inc. (
), and Staples, Inc. (
) are among the larger commercial partners to buy into the first generation deployment of Bloom Energy Servers. Bloom Energy installing 20-50 megawatts of capacity per year and has 100 megawatts of total installed capacity, presently. It currently handmakes its fuel cell systems at a pair of plants in Newark, New Jersey and Sunnyvale, Calif.
In November 2013 it
expanded to Japan
, where it's since scored several clients.
Today Bloom Energy is a bonafide power player with 83 patents and a growing installed base. It also has an abundance of cash having received $1.1B USD in capital funding. That total makes it one of the five biggest venture capital projects in history (not inflation adjusted). Its deep-pocketed backers include Kleiner Perkins Caufield & Byers (KPCB), Goldman Sachs Group Inc. (
), and Credit Suisse Group AG (
Bloom Energy founder and CEO K.R. Sridhar, Ph.D has snagged $1.1B USD in funding for his startup, but faces growing profitability concerns. [Image Source: AP]
Yet for all the positives, some fear that Bloom Energy may have fallen short on
its goal of achieving profitability
by 2013, as there was no official followup. Many
expected an initial public offering
(IPO) far sooner for the company;
suspicion of the company's bold financial promises
is on the rise.
Amid this uncertain backdrop, Redox Power believes it’s ready for its first serious commercial test in the wild. The startup is a spinoff from the
University of Maryland Energy Research Center (UMERC)
. Launched in Aug. 2013, the company continues to collaborate with the Univ. of Maryland.
Redox Power's founder,
Professor Eric Wachsman
, is an instructor at the university and is director of UMERC. He holds key patents on the technology which he claims will offer 100 times the density per cost of current cells, including Bloom's Energy Server. He claims his cells are 1/10th the cost of commercial alternatives and are also 1/10th the size.
Redox CEO Warren Citrin, left, and Professor Eric Wachsman believe they have the technology to blow away the Bloom Box. Now their effort -- 25 years in the making -- has been picked up by Microsoft for testing.
One strength of Redox Power's cell design is flexibility. It is designed to primarily run off natural gas, but can also generate power using propane, gasoline, biofuel, and hydrogen. At its maximum efficiency, when processing natural gas and doubling as backup heaters, the cells can output heat and electricity at 80 percent efficiency (and 70 percent efficiency for electrical generation only).
That's a good deal higher than Bloom Energy Servers, which are 60 percent efficient at optimal conditions.
II. Going Exotic
It took him 25 years to develop his unique brand of nanotechnology and microtechnology driven fuel cell, but Professor Waschman is finally seeing the payoff. Today he
holds many patents
on fuel cell technologies, but the asserted spec indicates that the technology is covered by
U.S. Patent No. 20110200910
, which was filed for in Oct. 2009 and granted in Aug. 2011.
Like Bloom Energy's "printed" fuel cell technology, this is a solid oxide fuel cell (SOFC) design. But from there they differ dramatically. Redox Power's cell clearly should enjoy a sizeable cost advantage based purely on the less expensive materials it uses. The big question is whether its novel design can overcome technical challenges of its less common design that is based on a kind of SOFC typically crippled by lower efficiencies.
Solid oxide fuel cells (SOFCs) are generally viewed as the most viable fuel cell technology. Upsides include their ability to be printed and their high efficiencies. Downsides include their tendency to be chemically poisoned by redox reactions and their use of expensive rare earth metals.
The first major difference in the Bloom Energy cell and the Redox Power cell is in the electrolyte. One crucial design flaw that reduces efficiency in most early SOFC designs is the loss of conductivity at lower temperatures (with room temperature being considered a very low temperature in fuel cell terms).
Bloom Energy is believed to use a relatively common yttria-stabilized zirconia (YSZ), but tries to solve the temperature problem by also adding a second printed electrolyte -- scandia-stabilized zirconia (ScSZ) [
U.S Patent No. 20080261099
]. ScSZ is much more efficient at conducting at lower temperatures than YSZ, but the downside is that scandium is very scarce and expensive.
Bloom Energy's "printed" fuel cells are about the same size as Redox Power's. They use yttrium and (possibly) scandium oxide electrolytes, where as Redox Power uses other metal oxides.
[Image Source: Getty Images]
Redox Power is believed to use cerium oxide (anode side) and bismuth oxide (cathode side). The bismuth oxide layer is thinner -- about 60 percent of the thick of the cerium oxide layer in production designs. The key problem with this kind of cell design is decomposition -- typically the cerium and bismuth oxides were both vulnerable to decomposition.
Redox Power's SOFCs (solid oxide fuel cells) use a dual electrolyte that's more affordable and more efficient than Bloom Energy.
The patent describes in some detail Professor Wachsman's solution -- to use a ceramic oxide of mid-size nanoparticles in an anion functional layer (AFL) that acts as a triple phase boundary enhancer to the cerium oxide electrolyte, which features smaller nanoparticle sizes. In layman's terms, by leveraging nanotechnology, the design is preserving the material's low temperature conductivity performance, while preventing degradation by separating the cerium oxide from the anode in a novel way.
Strategically structured nanoparticle deposits are key to prevent redox decomposition of Redox Power's attractive solid electrolyte. The technology is hard to duplicate given the careful formulation necessary. [Image Source: USPTO/Redox Energy]
There's a balance between performance and thickness of the bismuth oxide layer. Make it to thin and the cell underperforms. Make it too thick and the bismuth oxide layer will degrade based on interactions across the cerium oxide layer with the anode. Ultimately the patent indicates making the bismuth oxide layer roughly 60 percent of the cerium oxide layer in thickness offered the optimal balance.
Roughly 10 x 10 centimeters and a millimeter thick, Redox Power's fuel cells have a record-setting density of around 2 W/cm
For the interconnects between cells, Bloom Energy is believed to use a porous ceramic -- the standard approach. Redox Energy's patent details how it started with this kind of material, but was able to move to interconnects of stainless steel or other metals, thanks to its electrolyte and novel design. Prof. Waschman writes in the patent:
In addition to lower cost metal interconnects, the employment of lower temperatures allows the cell to be more tolerant of any thermal expansion mismatch, to be more easily sealed, to have less insulation, to consume less energy, have a more rapid startup, and to be more stable.
Redox Power's patent on technology shows it can achieve up to 2.5 watts of power, 6 amps/cm
, and 1 volt per cell [Image Source: USPTO/Redox Energy]
As a result of the design differences, Redox Power's cells operate at a temperature of around 650 °C, where as Bloom Energy's run at around 900 °C. The lower temperature accounts for part of Redox Power's efficiency edge.
III. Cost Kills the Bloom Box's Potential
Cost is another advantage for Redox Power. There's a modest cost savings in moving from ceramic to stainless steel interconnects, as mentioned. But perhaps the biggest savings is the switch to cheaper metal alloys in the metal oxide electrolytes.
Redox Power believes its cost are low enough to make it the first truly competitive large commercial fuel cell solution.
Not all rare metals are created equal. While most transition and rare Earth metals used in fuel cell production are scarce elements from ore deposits on the Earth's crust, some are rarer and more expensive to extract than others. Abundance aside, politics also play a key role in the cost and viability of rare Earth supplies. Some are almost exclusive controlled by China, who has leveraged its monopoly to drive up the price per kilogram. Other rare Earth elements and rare transitition metals enjoy a more competitor market with multiple producers worldwide. Those elements tend to be much more attractive as they cost less.
Unfortunately for Bloom Energy, it uses a combination of elements that are either very rare, or controlled by China. Hence the cost of printing its cells are likely relatively high, simply on the basis of its chosen electrolyte chemistry.
The biggest problem in its chemistry, cost-wise happens to also be its best performing electrolyte -- the scandium. Only 4 to 8 tons of scandium (a rare transition metal with properties similar to rare earth metals) is estimated to be produced annually,
Avalon Rare Metals Inc. (
). It's produced globally but in very small quantities as a coveted byproduct of metal production.
A reference page
from Jefferson Lab says that it is "too expensive for use."
A 2013 report
[PDF] from PFL Advisors, a consulting firm, says that prices have risen from around $500 USD per 99 percent pure kg of scandium metal to around $1,500 USD/kg.
Scandium production is extremely low and the rare metal's stock is almost exclusively controlled by China.
[Image Source: Wikimedia Commons]
Yttrium is mostly
extracted from clays
in China and is rather plentiful; production has skyrocketed from around 600 tons in 2001 to 8,900 tons in 2011 (of which 8,800 tons came from China),
U.S. Geological Survey
(USGS). Yttrium prices
a high of $180 USD/kg (99 percent pure) in Q3 2011. While they've since fallen to around , that's still pretty high. According to
the Strategic Metals & Rare Earth Letter
[PDF] from Feb. 2014, Yttrium has the highest shortage of any rare earth metal, clearly pointing to future price increases. Currently yttrium oxide sells for around
, while pure yttrium metal sells for around
Pacific Century Rare Earth Minerals, Ltd.
Yttrium is plentiful, but China controls production of the vital rare Earth metal, making it expensive.
[Image Source: Wikimedia Commons]
Redox Power's electrolyte chemistry is the payoff to the risk of redox fouling. From a cost standpoint it's dramatically better than Bloom Energy's for three reasons. First, it doesn't supplement its design with any excessively rare metal (e.g. scandium). Second, its metals are not control by China or some other monopolist. Third, since most SOFCs use YSZ as an electrolyte, Redox Power should have less competition for resources as it ramps up production.
The first crucial component of Redox Power's chemistry is the cerium oxide electrolyte. Cerium is also a rare earth element, but it's relatively abundant, with global annual production at 24,000 tons (
Minor Metals Trade Association
(MMTA)) -- more than yttrium, and much more than scandium. Extracted from the ores Bastnäsite and Monazite, it's also less sensitive to Chinese price manipulation as India, Brazil, and Australia are major producers.
The 2013 PFL Advisors report pegs the cost of cerium oxide at a relatively low
. Strong supply has dropped from a price of cerium oxide from nearly $20 USD/kg at its peak in 2011. Pure cerium (in metal ingot form) has a price of about
Cerium is produced worldwide in relative abundance, making it one of the less volatile rare Earths. [Image Source: Wikimedia Commons]
Bismuth production is around 3,000 tons annually, but a large amount of the production comes from Canada and Mexico, which again reduces vulnerability to Chinese price manipulation. It can be extracted either lead or copper ores, or be mined directly from crystal deposits. Bismuth
, shipping in 99 percent pure ingots.
Bismuth methal is known for its eye-catching maze-like crystal structure. It is produced in Canada and Mexico making it a geopolitically attractive compound. [Image Source: Wikimedia Commons]
As Chinese price manipulation has increased over the past few years, Bloom Energy has likely suffered -- which may be one reason why we haven't heard much from them, since the noisy launch in 2010. By contrast Redox Energy should be relatively immune to price spikes from resource rationing.
IV. Redox Power is Lean and Mean, Leaves Bloom Box Sucking Air
In volume and weight, Redox Power enjoys a sizeable lead over Bloom Energy, even if it isn't quite the ten-fold gain it initially claims.
Size and capacity-wise, currently Bloom's Energy Servers -- also referred to as "Bloom Boxes" are being offered in up to
200 KW-rated configurations
. The 200 KW configuration measures 26' 5" x 8' 7" x 6' 9" (feet and inches) (8.05 x 2.62 x 2.06 m). For reference sake, that's about the size of a shipping container. The system weighs 19.4 tons.
200 KW Bloom Energy servers (aka "Bloom Boxes") are roughly the size of a shipping container.
[Image Source: Univ. of Delaware]
Similar to Bloom Energy, Redox Power clusters its cells into a box-like design. The form name
, but is most frequently referred to by its nickname -- "The Cube." The Cube is much smaller than the Bloom Box (Bloom Energy Server).
A deconstructed Redox Power "Cube" fuel cell server
On its capacity and size, Redox Power writes:
Our first product is a 25 kW stationary power generation system fueled by natural gas. The Cube, as the device has been nicknamed, is about 1 m3 and under 1000 lbs and can be located in a variety of locations... [It] can comfortably power a gas station, moderately sized grocery store or small shopping plaza.
Additional power offerings will follow. Using different-sized fuel cell stacks, the company can offer The Cube at 5 kW, to provide always-on electricity for an average American home, or up to 80 kW in one system. Cubes can also be combined to provide even more power in a modular fashion.
Redox Power's "Cube" Server is about the size of a refrigerator, but pumps out 25 KW.
So Redox Power's Cube is about 1x1x1 meter (3.28 x 3.28 x 3.28 feet) and produces 25 KW. Based on a crude weight estimate (~900 lb) we can guess that it weighs about 0.4 metric tons (it may actually weigh less).
A Redox Power Cube packs about 4 times the power per volume as the Bloom Box. It also costs less to make, is nearly 10 percent more efficient (on paper), and weighs 50 percent less.
If you stacked the Redox Power systems tightly, roughly 32 would fit into the footprint of the 200 KW Bloom Energy Server. That's 800 KW, which gives a crude estimate that Redox packs
four times the power
into the same volume as Bloom Energy. That makes a huge difference in portability, installation costs, and serviceability.
The same napkin math gives you a rough estimate of the weight of that combined 800 KW stack -- 12.8 tons. That's roughly
50 percent less weight per space
than Bloom Energy's system. Again that saves on a variety of expenses. That all works out to about
a six-fold gain in true density (volume/weight)
, which is pretty impressive.
V. Microsoft is Excited to Have the Market's Best Fuel Cells
The data center is a pivotal pillar of Microsoft's business model. With products such as
, and the Xbox One (which
features cloud-boosted processing
), Microsoft is increasingly making cloud storage and processing a central pieces of its consumer and enterprise product line.
new CEO Satya Nadella
was the company's chief cloud evangelist, so this direction is not surprising. But the direction places pressure on Microsoft to up its game in terms of data center reliability, hardware costs, pollution mitigation, and energy efficiency. If Microsoft cannot keep up with
rivals like Google
) and Facebook, Inc. (
) who are
masters at data center construction
, its vision will fail.
New Microsoft CEO Satya Nadella's vision hinges on data center-driven cloud computing.
[Image Source: Reuters]
In that regard the Redox Power Cube could play a key role in the software giant's long-term plans. Microsoft plans to directly install "The Cube" on site at its data centers, replacing its current stable of
expensive, polluting, and noisy diesel generators
. This is actually a big deal as Microsoft's current diesel generators produce so much pollution that neighboring towns are
complaining of higher rates of respiratory illnesses
similar to the kind seen in
smog-laden Chinese cities
Bloom Boxes are generally regarded as money losers in cost analysis, which do not account for backlash regarding air pollution. But thanks to dramatic space and cost savings of the Redox Power Cubes, Microsoft's first deployment may actually turn a profit, even before air pollution's "soft costs" are considered.
A Microsoft data center [Image Source: Microsoft]
And Microsoft has a grant to cover most of its initial deployment costs, so this is a pretty exciting opportunity for the cloud computing giant.
, Senior Research Program Manager for Microsoft Global Foundation Service, comments on the collaboration:
Our vision is to bring the power plant directly into the datacenter by integrating fuel cell stacks into every server cabinet, effectively eliminating energy loss that otherwise occurs in the energy supply chain and doubling the efficiency of traditional datacenters.
While Redox Power is perhaps Microsoft's best shot at a cost effective alternative to its noxious fossil fuel generators, Microsoft is also working on a number of other alternative energy ideas for its data centers.
A joint effort between Microsoft, FuelCell Energy, and the
University of Wyoming
– co-funded by Microsoft and the state of Wyoming -- is exploring
using wastewater (feces-rich water from sewage) as a source natural gas for fuel cells
. The FuelCell energy system Microsoft is testing is a 300 KW stack, roughly the size of a shipping container. It's comparable to the Bloom Energy Server in most regards, but likely inferior to the new Redox Power system.
A similar 300 kW stack from FuelCell Energy is deployed on the campus of Yale University.
[Image Source: Yale University]
In Feb. 2014, Microsoft showed off an all-in-one self-contained server rack and fuel cell power system, in collaboration with the
National Fuel Cell Research Center
(UCI). The partner for that design and finer performance details were not generally available, but it's likely that Microsoft will look to leverage those ideas in its partnership with Redox Power.
Microsoft has also
signed a three-year agreement
University of Texas at San Antonio
(UTSA) to research green data center solutions.
(All images courtesy of Redox Power unless otherwise noted.)
Microsoft [MSDN Blogs]
Univ. of Maryland [background]
This article is over a month old, voting and posting comments is disabled
6/26/2014 3:20:11 PM
"using wastewater (feces-rich water from sewage)"
Wonder when I can get one that site on top of my septic tank? Just imagine asking everyone to poop more to keep the lights on!
"This week I got an iPhone. This weekend I got four chargers so I can keep it charged everywhere I go and a land line so I can actually make phone calls." -- Facebook CEO Mark Zuckerberg
Microsoft Boosts Free OneDrive Storage to 15GB, Office 365 Users to Get 1TB
June 23, 2014, 12:06 PM
Microsoft Wants Quick Death for Office 2003, Leads Users to Office 365
March 25, 2014, 9:02 AM
Meet Microsoft's New CEO: Satya Nadella
February 4, 2014, 9:04 AM
San Francisco to Google: Barge Out
February 3, 2014, 6:52 PM
Chinese Media Puts Positive Spin on Its Smog Problem, Touts 5 "Benefits"
December 11, 2013, 12:39 PM
Samsung Adds 2 TB 850 EVO, PRO SSDs for $800, $1000
July 7, 2015, 4:23 PM
Seagate Senior Researcher: Heat Can Kill Data on Stored SSDs
May 13, 2015, 2:49 PM
How to Recover Most Apps After Your NVIDIA Driver Crashes in Windows 10
March 30, 2015, 12:54 PM
Tinkerer Gets Old School Mac Plus Running on the Modern Web
March 24, 2015, 6:41 PM
Facebook-Backed Oculus Rift's Release Date Slips to 2016; Valve and HTC Salivate
March 16, 2015, 5:58 PM
Hackers Steal Roughly $1 Billion From Banks Using Malware RAT
February 17, 2015, 9:30 AM
Most Popular Articles
Exclusive: If Intel and Micron's "Xpoint" is 3D Phase Change Memory, Boy Did They Patent It
July 29, 2015, 10:52 PM
Quick Note: Apple Watch to Get Brick and Mortar Boost From Best Buy
July 27, 2015, 3:00 PM
Windows 10 to Get New Features in October Service Release 2 (SR2)
July 30, 2015, 5:50 PM
As iPad Sales Wane and Watch Flops, iPhone Saves Apple's Profit With Its Heroics
July 22, 2015, 6:13 PM
Editorial: Reddit Allows Itself to be Hijacked as a Hate Platform For Racist Bigots
July 21, 2015, 6:32 PM
Latest Blog Posts
Sceptre Airs 27", 120 Hz. 1080p Monitor/HDTV w/ 5 ms Response Time for $220
Dec 3, 2014, 10:32 PM
Costco Gives Employees Thanksgiving Off; Wal-Mart Leads "Black Thursday" Charge
Oct 29, 2014, 9:57 PM
"Bear Selfies" Fad Could Turn Deadly, Warn Nevada Wildlife Officials
Oct 28, 2014, 12:00 PM
The Surface Mini That Was Never Released Gets "Hands On" Treatment
Sep 26, 2014, 8:22 AM
ISIS Imposes Ban on Teaching Evolution in Iraq
Sep 17, 2014, 5:22 PM
More Blog Posts
Copyright 2015 DailyTech LLC. -
Terms, Conditions & Privacy Information